search

ksahlin / strobealign

Aligns short reads using dynamic seed size with strobemers
MIT License
144 stars 17 forks source link

Supplementary reads #13

Open kcleal opened 2 years ago

kcleal commented 2 years ago

Hi, Just a question - does strobealign support split-reads / supplementary mappings with a flag value of 2048? Thanks Kez

ksahlin commented 2 years ago

Hi @kcleal,

Not yet, but I have found that it is possible to report supplementary alignments semi-easily with the current code structure. Release 0.7 will support supplementary alignments. Will keep this issue open until it is implemented.

Release 0.6 will come in a couple of days with further improvements.

kcleal commented 2 years ago

Thanks!

Supplementary mappings are really handy for calling SVs, so would be great if you can implement this

ksahlin commented 2 years ago

Note to developer: When implementing this, consider the runtime overhead. It could be modularly implemented,i.e., invoked with a parameter to strobealign and enter a 'supplementary alignment check'. This can probably be inferred from the vector containing secondary alignments - but need to keep track/pair up aligned coordinates from the different alignment structs.

marcelm commented 2 years ago

I realized just now that our tiny test genome (phiX) is circular and therefore good for testing supplementary alignments: Some of the phiX reads "wrap around" the FASTA reference, that is, a prefix aligns to the 3' end of the reference and a suffix aligns to its 5' end. This would need to be modelled with supplementary alignments.

edg1983 commented 2 years ago

Supplementary reads are not yet implemented in the last release (0.7.1) right? I'm testing it to see how strobealign perform in our current WGS pipeline and I see no supplementary reads. Having supp/split reads is crucial for many SV callers, like Manta and Lumpy that we use in our detection workflow.

marcelm commented 2 years ago

Yes, this is not implemented, but is definitely on the agenda.

ksahlin commented 2 years ago

Yes @edg1983, I was a bit hasty promising this in v0.7. Supplementary alignments is one of our highest priority features to implement though. So hopefully we will come back to this soon.

kcleal commented 1 year ago

Hi @kcleal, any ideas when this might be added? I would like to start using strobealign for SV analysis. Thanks!

ksahlin commented 1 year ago

Hi @kcleal,

Thanks for the interest! What I can say is that @marcelm is currently working on improving/speeding up extension alignment, and it is slightly related to supplementary alignments. Also, it is definitely in our interest to add such feature. But I don't want to interfere with his todo list, so I will simply pass this question to @marcelm.

To Marcel: I guess we also need to assess if it fits the number of hours left on the project and the priorities.